Identifying supply chain credit risk in China’s textile industry using an ensemble explainable artificial intelligence model
Skip to main content
- This study develops a novel interpretable hybrid XAI–PXL model that integrates PSO-weighted parallel XGBoost and Logistic regression to balance predictive performance and transparency for supply chain financial credit risk identification.
- A multi-layer feature screening workflow eliminates redundant indicators via missing value, IV, correlation and PSI tests.
- Out-of-sample comparative validation confirms the XAI–PXL composite model outperforms standalone XGBoost and Logistic regression, achieving the highest balanced metrics for corporate default risk detection.
The development of artificial intelligence and financial technology has introduced new growth opportunities as well as creating complex challenges in credit risk identification, communication and control for supply chain enterprises. To address these, this study focuses on data mining of financial credit risk information and knowledge in the supply chain from the perspective of explainable artificial intelligence (XAI). By applying scientific and quantitative research methods we develop an XAIPSO– XGBoost-Logistic (XAI–PXL) model, which combines XAI using the particle swarm optimization, extreme gradient boosting and logistic regression algorithms. Our research shows that the combined XAI–PXL model has a good risk identification and classification performance and it outperforms the individual XGBoost and logistic regression algorithms in identifying credit risks in supply chain finance. The explanatory power of the model is analyzed using the Shapley additive explanations method and the order of explanatory power obtained for the characteristic variables input into the model. This study contributes to the literature on risk identification from the perspective of XAI, demonstrating strong theoretical and practical significance in identifying supply chain financial credit risks, acquiring supply chain risk knowledge, supporting decision-making for governments and enterprises.
Copyright Infopro Digital Limited. All rights reserved.
You may share this content using our article tools. As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (clause 2.4), an Authorised User may only make one copy of the materials for their own personal use. You must also comply with the restrictions in clause 2.5.
If you would like to purchase additional rights please email info@risk.net
You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.
